Effects of high intensity interval exercise on cerebrovascular function: A systematic review

High intensity interval exercise (HIIE) improves aerobic fitness with decreased exercise time compared to moderate continuous exercise. A gap in knowledge exists regarding the effects of HIIE on cerebrovascular function such as cerebral blood velocity and autoregulation. The objective of this systematic review was to ascertain the effect of HIIE on cerebrovascular function in healthy individuals. We searched PubMed and the Cumulative Index to Nursing and Allied Health Literature databases with apriori key words. We followed the Preferred Reporting Items for Systematic Reviews. Twenty articles were screened and thirteen articles were excluded due to not meeting the apriori inclusion criteria. Seven articles were reviewed via the modified Sackett’s quality evaluation. Outcomes included middle cerebral artery blood velocity (MCAv) (n = 4), dynamic cerebral autoregulation (dCA) (n = 2), cerebral de/oxygenated hemoglobin (n = 2), cerebrovascular reactivity to carbon dioxide (CO2) (n = 2) and cerebrovascular conductance/resistance index (n = 1). Quality review was moderate with 3/7 to 5/7 quality criteria met. HIIE acutely lowered exercise MCAv compared to moderate intensity. HIIE decreased dCA phase following acute and chronic exercise compared to rest. HIIE acutely increased de/oxygenated hemoglobin compared to rest. HIIE acutely decreased cerebrovascular reactivity to higher CO2 compared to rest and moderate intensity. The acute and chronic effects of HIIE on cerebrovascular function vary depending on the outcomes measured. Therefore, future research is needed to confirm the effects of HIIE on cerebrovascular function in healthy individuals and better understand the effects in individuals with chronic conditions. In order to conduct rigorous systematic reviews in the future, we recommend assessing MCAv, dCA and CO2 reactivity during and post HIIE.


Introduction
High intensity interval exercise (HIIE) has emerged at the forefront of exercise regimens due to the shorter activity time needed to benefit [1][2][3]. HIIE confers similar or significant increased aerobic fitness compared to conventional moderate intensity continuous exercise [1,[4][5][6][7]. While aerobic fitness is a measure of increased cardiovascular health, the entire vascular system (including the cerebral vascular system) may be improved following increased aerobic fitness [8]. With aging, higher aerobic fitness is associated with a lower risk of stroke and dementia [9,10]. A review and meta-analysis of HIIE in healthy adults has shown significant increases in aerobic fitness [1,5,6,11]. Preliminary evidence has also shown HIIE may improve cognitive function [12]. However, the effects of HIIE on cerebrovascular function have not been systematically reviewed.
Cerebrovascular function is the ability of the cerebral blood vessels to deliver oxygen and nutrients for neuronal metabolism and maintain cerebral blood flow through dynamic autoregulation (dCA). dCA is the ability of the brain to sustain a constant cerebral blood flow despite large fluctuations in peripheral blood pressure [13,14]. During resting conditions, cerebral blood flow responds to arterial blood pressure fluctuations, neuronal metabolism, cortical activation, arterial blood gases and cardiac output [15]. Cerebral blood flow can be measured at rest using magnetic resonance imaging or transcranial Doppler ultrasound (TCD). Middle cerebral artery blood velocity (MCAv) measured by TCD is the only technique to measure cerebral blood flow during exercise, with high temporal resolution [16]. MCAv is linearly related to cerebral blood flow with the caveat that the MCA diameter remains unchanged [17].
A normal cerebrovascular response to submaximal moderate continuous exercise results in increased MCAv [18][19][20], increased cerebral oxygenation [21,22] and sustained dCA [23,24]. MCAv has been shown to concomitantly increase as exercise intensity increases, up to moderate intensity [15,18,[25][26][27][28]. MCAv is affected differently during high intensity exercise. During continuous high intensity exercise and hyperventilation, MCAv is decreased due to a reduction in arterial carbon dioxide (CO 2 ) [29, 30] causing downstream arteriole constriction [15,31]. Cerebrovascular reactivity is the ability of the small vessels in the brain to vasodilate and vasoconstrict in response to fluctuating CO 2 levels [32,33]. The cerebrovascular response to HIIE may differ from continuous high intensity exercise due to the repetitive short interval bouts that rapidly increase blood pressure which may cause cerebrovascular hyper-perfusion [34,35]. If neuroprotective mechanisms of the brain, such as dCA, do not respond quickly to the repetitive and rapid increases in blood pressure, HIIE could elevate the risk for leakage within the blood brain barrier [34,36]. For clinical populations with cerebrovascular impairment, such as stroke [37][38][39], the cerebrovascular response to HIIE may play an important role in guiding exercise prescription [36].
Previous scientific statements and narrative reviews have recounted the molecular, hemodynamic and structural processes (i.e. CO 2 , nitric oxide, systemic blood pressure, vessel compliance, glial cell integrity) associated with the cerebrovascular response that may occur during HIIE [36,40]. However, these detailed narrative reviews [41, 42] did not report the statistical findings of previous studies showing cerebrovascular function during HIIE. To our knowledge, our current systematic review is the first to systematically search and report the results of the dynamic cerebrovascular response during HIIE. Reporting the cerebrovascular response during HIIE is important because it provides objective results to support the previously described narrative statements on hemodynamic processes during HIIE [41,42]. The purpose of this systematic review was to address the gap in knowledge and report the various study results of HIIE on cerebrovascular function compared to moderate continuous exercise or rest conditions. We systematically examined the results of HIIE studies in healthy individuals based on the operationalization of cerebrovascular outcomes.

Methods
This review follows the guidelines for Preferred Reporting Items for Systematic Reviews [43]. Literature searches and reviews were performed using PubMed and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) databases. The University of Kansas Medical Center Online Library system was used to access these databases in February, March, and June 2020. In this systematic review, we included peer-reviewed manuscripts written in English from January 2010 to June 2020.
Key words used to search the databases included "high intensity interval training", "HIIT", "high intensity interval exercise", "HIIE" AND "cerebral blood flow", "cerebral blood velocity", "dynamic autoregulation". We believe these key words primarily reflect the high intensity interval intervention and cerebrovascular function outcome measures. The main outcomes of this systematic review were MCAv, an indirect measure of cerebral blood flow, and dCA, a measure of cerebrovascular homeostasis during peripheral blood pressure changes [30,[44][45][46][47][48]. MCAv supplies oxygen and nutrients to neurons while dCA maintains stable perfusion [49]. However, additional cerebrovascular measures were also included such as oxygenated hemoglobin [50-52], cerebrovascular reactivity [46,53], cerebrovascular conductance index and cerebrovascular resistance index [45,54]. Oxygenated hemoglobin is an important measure of aerobic metabolism within cerebral tissue using near-infrared spectrometry [55]. Cerebrovascular reactivity is a measure of cerebrovascular regulation [56] and shows the ability of the vessels to vasodilate or vasoconstrict to a stimulus [57]. Cerebrovascular conductance index is a measure of the conductance of peripheral blood pressure to cerebral blood velocity and is calculated as MCAv/mean arterial pressure (MAP) [45]. Cerebrovascular resistance index (MAP/MCAv) measures the resistance of cerebral perfusion pressure to cerebral blood velocity [45].
The identified abstracts from PubMed and CINAHL were screened using the following inclusion criteria: 1) experimental or quasi-experimental, 2) aerobic exercise identified as the primary means of performing HIIE, 3) cerebrovascular measures were primary or secondary outcomes and 4) human subjects across the lifespan with no current disease. After the removal of duplicates, two researchers screened titles/abstracts for inclusion criteria (A.W. and M.A.). The full texts were examined, and data extracted (A.W. and M.A.). If the authors were unable to come to an agreement, a third author moderated incongruity (A.F.).
A quality review was performed for each article using the modified version of Sackett's 1981 criteria [58]. We critically analyzed each article's study design, population, HIIE protocols, cerebrovascular outcomes and results. If an article did not report enough information to determine sufficient quality criteria a "No" rating was given. Articles were rated based on the level of evidence including level I for large randomized control trials, level II for small randomized trial, level III for nonrandomized design, Level IV for case series and Level V for case reports [59].

Results
The search methods resulted in 67 articles. After removal of duplicates, 15 [45, 46, 50, 53, 60-69] articles were identified in PubMed and 5 [47, 51, [70][71][72] new articles in CINAHL. During the initial screening of titles/abstracts, 11 articles were excluded due to HIIE not being the primary experimental protocol performed (n = 6), studies not measuring cerebral arteries (n = 4) and an animal study (n = 1). Studies that combined other interventions with HIIE were excluded due to the confounding variables that could affect cerebrovascular outcomes (see S1 Table). After the full text assessment, two articles were excluded due to not meeting experimental or quasi experimental criteria (n = 2). See Fig 1 for flow diagram of article selection.
We included seven articles describing cerebrovascular outcomes following HIIE within this review [44-47, 50, 51, 53]. The full texts are described in Table 1. Of the articles reviewed, six were small, randomized trials and one nonrandomized cross-over trial. All the studies   A constant between all studies included an active (rather than passive) recovery interval between sprints. However, the intensity and duration of recovery intervals differed greatly. Cerebrovascular reactivity. Cerebrovascular reactivity to CO 2 were reported during a single bout of HIIE (n = 1) [53] and during a 12-week HIIE intervention (n = 1) [46]. After a single bout of HIIE, cerebrovascular reactivity to higher CO 2, or hypercapnia, was significantly decreased by 37% and remained an hour later [53]. The reduced cerebrovascular reactivity to hypercapnia was also significantly different than moderate intensity and control. Cerebrovascular reactivity to lower CO 2 , or hypocapnia, was not significantly different following a single bout of HIIE [53]. Cerebrovascular reactivity to CO 2 was also not significantly different following 12 weeks of HIIE [46]. Cerebrovascular conductance and resistance. Cerebrovascular conductance index and cerebrovascular resistance index were only reported in a single study [45]. A 6-week HIIE intervention reported no significant changes in cerebrovascular conductance index or cerebrovascular resistance index [45].

Quality review
The quality review of each study is presented in

Discussion
This review met the objective of reporting the results of various HIIE studies and the effects on operationalized cerebrovascular function in healthy individuals. This review is the first to report the effects HIIE on cerebrovascular function compared to moderate continuous exercise and rest in healthy individuals. In general, we found that the acute and chronic effects of HIIE on cerebrovascular function vary largely depending on the methods and outcomes measured.

MCAv
In these studies, 6-to 12-week HIIE interventions had no effect on resting MCAv in healthy individuals. No significant change in resting MCAv may be due to the HIIE intervention duration being too short. Also a ceiling effect may be observed for young, healthy individuals and could explain no changes in resting MCAv [19]. During a single bout of HIIE, hyperventilation and downstream arteriole vasoconstriction may explain the acute decreases in exercise MCAv compared to moderate continuous exercise [15,31,73,74]. Vasoconstriction may play a protective role during HIIE due to heightened peripheral blood pressure potentially causing hyper-perfusion [75] [76]. Therefore, exercise MCAv should be reported for each interval of HIIE rather than an average of the entire exercise bout.
dCA After a 6-week intervention and single bout of HIIE, dCA phase was decreased compared to rest. The chronic effects of HIIE on dCA phase may be due to elevated cardiorespiratory fitness in endurance trained individuals being associated with attenuated dCA [45,48]. In healthy individuals, increased frequency within MCAv and MAP waveforms (that can occur with HIIE) may cause a reduction in phase due to dCA being a high-pass filter [77,78]. Burma et al.
[44] also suggests that systolic phase may reveal greater changes in dCA than both diastolic and mean phase. After a single bout of HIIE, reduction in systolic phase extended up to 4 hours and therefore the common approach of abstaining from exercise 12 hours before research studies [79][80][81] may be too conservative [44].
Although not included in this review due to the observational study design, contradictory evidence of sustained dCA during HIIE has been reported [63]. Differences in exercise parameters between HIIE may be the cause to contradictory findings due to exhaustive exercise showing decreased dCA [34,82]. More studies are needed to confirm the acute and chronic decreases in dCA following HIIE.

De/oxygenated hemoglobin
After a 16-week HIIE intervention, oxygenated and deoxygenated hemoglobin during cortical activation did not change [50]. However, the 16-week HIIE intervention decreased reaction time during cortical activation and therefore may have increased efficiency of cortical oxygen use [50]. During a single bout of HIIE oxygenated and deoxygenated hemoglobin increased compared to moderate continuous exercise [51]. As suggested by Coetsee et al, increased oxygenated hemoglobin during neuronal activation may suggest engaging additional regions of the brain [50]; while decreased oxygenated hemoglobin may indicate reduced neuronal activity due to task-efficiency [50]. The acute and chronic effects of HIIE on oxygenated and deoxygenated hemoglobin still needs further investigation due to each only being reported in a single study. A 12-week HIIE intervention did not significantly change cerebrovascular reactivity which could be due to vascular desensitization from chronic exposure to CO 2 during HIIE [36,83]. Following a single bout of HIIE, cerebrovascular reactivity to hypercapnia was decreased showing the inability of the cerebrovascular system to maximally vasodilate. The maximal capacity for vasodilation after HIIE may be reduced following HIIE due to prolonged cerebrovascular vasoconstriction that occurs with hyperventilation during HIIE [36,53]. Cerebrovascular reactivity to hypocapnia was not changed following a single bout of HIIE due to the ability of the vessels to vasoconstrict remaining intact [53]. The reduction in cerebrovascular reactivity to higher CO 2 remains an hour after HIIE. Therefore, the authors conclude again that the common approach of abstaining from exercise 12 hours before research studies [54,80,84] may be too conservative [53].

Cerebrovascular conductance and resistance
Cerebrovascular conductance and resistance were not significantly changed following a 6-week HIIE intervention. While a 6-week HIIE intervention significantly improved peripheral arterial conductance and resistance [4], this change in the peripheral arteries may not be demonstrated in the cerebrovascular arterial conductance or resistance [41,45]. However, due to the cerebrovascular conductance or resistance index being reported in only a single study, no conclusive effects of HIIE can be determined.

Future research
We recommend future research on the effects of HIIE on cerebrovascular function should include: 1) examining the cerebrovascular response during HIIE before and after an intervention of HIIE, 2) analyzing cerebrovascular outcomes during each separate interval of HIIE rather than an average of the entire bout, 3) simultaneously measuring MCAv, blood pressure, heart rate, and CO 2 during HIIE, 4) measuring cerebrovascular outcomes during HIIE, immediately following HIIE and at a follow up 30 minutes to 4 hours post exercise.

Limitations
The authors acknowledge a risk of publication bias by only including peer-reviewed articles written in English and did not include grey literature. The cerebrovascular function measures included within this review vary greatly and have vast heterogeneity. The overall quality of studies is moderately poor due to the lack of avoiding contamination, not blinding the assessment, and scarce reporting of the reliability and validity of the outcomes measured. These studies report the effects of HIIE on cerebrovascular function in healthy young individuals which limits generalizability and cannot be translated to clinical populations with altered cerebrovascular function at baseline, such as stroke [37,39,85,86].
While HIIE is not a new mode of exercise, studying cerebrovascular measures during HIIE is novel. There are potential limitations to using TCD during HIIE and MCAv may be underestimated [87]. Cerebral oxygenation may also be underestimated due to the two-channel near infrared spectrometer not measuring the motor, occipital, or parietal cortex [50,51]. Authors could only identify seven small studies with the oldest article dating back to 2015. The primary outcome of MCAv (n = 4) and dCA (n = 2) were reported in few studies with low power. Therefore, a meta-analysis could not be performed due to insufficient mathematical combination.

Conclusion
This review has provided preliminary information studying the effects of HIIE on cerebrovascular function. Currently, there are a sparse number of research studies with moderately poor quality criteria that have reported the acute and chronic effects of HIIE on cerebrovascular function. An increased amount of studies and greater quality of research avoiding contamination, blinding the assessments, and reporting reliability and validity is needed. Randomized controlled trials with large sample sizes are needed to conduct a meta-analysis to combine and statistically analyze the summary results of HIIE on cerebrovascular function. Additionally, more studies are needed to determine the optimal interval parameters of HIIE to provide a consistent exercise dose between studies.
With increased interest in healthy brain aging and implementing interventions to maintain or improve brain health [88], studying the effects of HIIE interventions are critically needed [41,42]. While this review only included healthy individuals, we provide an early reference to understanding "normal" physiological effects of HIIE on cerebrovascular function and the need to compare to clinical populations. Researchers should make further efforts to investigate and report the effects of HIIE on diverse measures of cerebrovascular function. To do so, it is imperative that researchers implement high quality criteria within the planning of future studies.